[apple/icu.git] / icuSources / data / cldr-icu-readme.txt

# Copyright (C) 2016 and later: Unicode, Inc. and others.
# License & terms of use: http://www.unicode.org/copyright.html
# Copyright (C) 2010-2014, International Business Machines Corporation and others.
# All Rights Reserved.                  
#
# Commands for regenerating ICU4C locale data (.txt files) from CLDR.
#
# The process requires local copies of
#    - CLDR (the source of most of the data, and some Java tools)
#    - ICU4J  (used only for checking the converted data)
#    - ICU4C  (the destination for the new data, and the source for some of it)
#             (Either check out ICU4C from Subversion, or download the additional 
#              icu4c-*-data.zip file so that the icu/source/data/ directory is fully
#              populated.)
#
# For an official CLDR data integration into ICU, these should be clean, freshly
# checked-out. For released CLDR sources, an alternative to checking out sources
# for a given version is downloading the zipped sources for the common (core.zip)
# and tools (tools.zip) directory subtrees from the Data column in
# [http://cldr.unicode.org/index/downloads].
#
# The versions of each of these must match. Included with the release notes for
# ICU is the version number and/or a CLDR svn tag name for the revision of CLDR
# that was the source of the data for that release of ICU.
#
# Note: Some versions of the OpenJDK will not build the CLDR java utilities.
#   If you see compilation errors complaining about type incompatibilities with
#   functions on generic classes, try switching to the Sun JDK.
#
# Besides a standard JDK, the process also requires ant
# (http://ant.apache.org/),
# plus the xml-apis.jar from the Apache xalan package
# (http://xml.apache.org/xalan-j/downloads.html).
#
# Note: Enough things can (and will) fail in this process that it is best to
#   run the commands separately from an interactive shell. They should all
#   copy and paste without problems.
#
# It is often useful to save logs of the output of many of the steps in this
# process. The commands below put log files in /tmp; you may want to put them
# somewhere else.
#
#----
#
# There are several environment variables that need to be defined.
#
# a) Java- and ant-related variables
#
# JAVA_HOME:     Path to JDK (a directory, containing e.g. bin/java, bin/javac,
#                etc.); on many systems this can be set using
#                `/usr/libexec/java_home`.
#
# ANT_OPTS:      You may want to set:
#
#                -Xmx3072m, to give Java more memory; otherwise it may run out
#                 of heap.
#
# b) CLDR-related variables
#
# CLDR_DIR:      Path to root of CLDR sources, below which are the common and
#                tools directories.
# CLDR_CLASSES:  Defined relative to CLDR_DIR. It only needs to be set if you
#                are not running ant jar for CLDR and have a non-default output
#                folder for cldr-tools classes.
#
# c) ICU-related variables
# These variables only need to be set if you're directly reusing the
# commands below.
#
# ICU4C_DIR:     Path to root of ICU4C sources, below which is the source dir.
#
# ICU4J_ROOT:    Path to root of ICU4J sources, below which is the main dir.
#
#----
#
# If you are adding or removing locales, or specific kinds of locale data,
# there are some xml files in the ICU sources that need to be updated (these xml
# files are used in addition to the CLDR files as inputs to the CLDR data build
# process for ICU):
#
#    icu/trunk/source/data/icu-config.xml - Update <locales> to add or remove
#                CLDR locales for inclusion in ICU. Update <paths> to prefer
#                alt forms for certain paths, or to exclude certain paths; note
#                that <paths> items can only have draft or alt attributes.
#
#                Note that if a language-only locale (e.g. "de") is included in
#                <locales>, then all region sublocales for that language that
#                are present in CLDR data (e.g. "de_AT", "de_BE", "de_CH", etc.)
#                should also be included in <locales>, per PMC policy decision
#                2012-05-02 (see http://bugs.icu-project.org/trac/ticket/9298).
#
#    icu/trunk/source/data/build.xml - If you are adding or removing break
#                iterators, you need to update  <fileset id="brkitr" ...> under
#                <target name="clean" ...> to clean the correct set of files.
#
#    icu/trunk/source/data/xml/      - If you are adding a new locale, break
#                iterator, collation tailoring, or rule-based number formatter,
#                you may need to add a corresponding xml file in (respectively)
#                the main/, brkitr/, collation/, or rbnf/ subdirectory here.
#
#----
#
# For an official CLDR data integration into ICU, there are some additional
# considerations:
#
# a) Don't commit anything in ICU sources (and possibly any changes in CLDR
#    sources, depending on their nature) until you have finished testing and
#    resolving build issues and test failures for both ICU4C and ICU4J.
#
# b) There are version numbers that may need manual updating in CLDR (other
#    version numbers get updated automatically, based on these):
#
#    common/dtd/ldml.dtd                            - update cldrVersion
#    common/dtd/ldmlBCP47.dtd                       - update cldrVersion
#    common/dtd/ldmlSupplemental.dtd                - update cldrVersion
#    tools/java/org/unicode/cldr/util/CLDRFile.java - update GEN_VERSION
#
# c) After everything is committed, you will need to tag the CLDR, ICU4J, and
#    ICU4C sources that ended up being used for the integration; see step 17
#    below.
#
################################################################################

# 1a. Java and ant variables, adjust for your system

export JAVA_HOME=`/usr/libexec/java_home`
export ANT_OPTS="-Xmx3072m"

# 1b. CLDR variables, adjust for your setup; with cygwin it might be e.g.
# CLDR_DIR=`cygpath -wp /build/cldr`

export CLDR_DIR=$HOME/cldr/trunk
#export CLDR_CLASSES=$CLDR_DIR/tools/java/classes

# 1c. ICU variables

export ICU4C_DIR=$HOME/icu/trunk/icu4c
export ICU4J_ROOT=$HOME/icu/trunk/icu4j

# 2. Build the CLDR Java tools

cd $CLDR_DIR/tools/java
ant jar

# 3. Configure ICU4C, build and test without new data first, to verify that
# there are no pre-existing errors (configure shown here for MacOSX, adjust
# for your platform).

cd $ICU4C_DIR/source
./runConfigureICU MacOSX
make all 2>&1 | tee /tmp/icu4c-oldData-makeAll.txt
make check 2>&1 | tee /tmp/icu4c-oldData-makeCheck.txt

# 4. Build the new ICU4C data files; these include .txt files and .mk files.
# These new files will replace whatever was already present in the ICU4C sources.
# This process uses ant with ICU's data/build.xml and data/icu-config.xml to
# operate (via CLDR's ant/CLDRConverterTool.java and ant/CLDRBuild.java) the
# necessary CLDR tools including LDML2ICUConverter, ConvertTransforms, etc.
# This process will take several minutes.
# Keep a log so you can investigate anything that looks suspicious.

cd $ICU4C_DIR/source/data
ant clean
ant all 2>&1 | tee /tmp/cldr-newData-buildLog.txt

# 5. Check which data files have modifications, which have been added or removed
# (if there are no changes, you may not need to proceed further). Make sure the
# list seems reasonable.

svn status

# 6. Fix any errors, investigate any warnings. Some warnings are expected,
# including  warnings for missing versions in locale names which specify some
# collationvariants, e.g.
#   [cldr-build] WARNING (ja_JP_TRADITIONAL): No version #??
#   [cldr-build] WARNING (zh_TW_STROKE): No version #??
# and warnings for some empty collation bundles, e.g.
#   [cldr-build] WARNING (en):  warning: No collations found. Bundle will ...
#   [cldr-build] WARNING (to):  warning: No collations found. Bundle will ...
#
# Fixing may entail modifying CLDR source data or tools - for example,
# updating the validSubLocales for collation data (file a bug if appropriate).
# Repeat steps 4-5 until there are no build errors and no unexpected
# warnings.

# 7. Now rebuild ICU4C with the new data and run make check tests.
# Again, keep a log so you can investigate the errors.

cd $ICU4C_DIR/source
make check 2>&1 | tee /tmp/icu4c-newData-makeCheck.txt

# 8. Investigate each test case failure. The first run processing new CLDR data
# from the Survey Tool can result in thousands of failures (in many cases, one
# CLDR data fix can resolve hundreds of test failures). If the error is caused
# by bad CLDR data, then file a CLDR bug, fix the data, and regenerate from
# step 4. If the data is OK but the testcase needs to be updated because the
# data has legitimately changed, then update the testcase. You will check in
# the updated testcases along with the new ICU data at the end of this process.
# Note that if the new data has any differences in structure, you will have to
# update test/testdata/structLocale.txt or /tsutil/cldrtest/TestLocaleStructure
# may fail.
# Repeat steps 4-7 until there are no errors.

# 9. Now run the make check tests in exhaustive mode:

cd $ICU4C_DIR/source
export INTLTEST_OPTS="-e"
export CINTLTST_OPTS="-e"
make check 2>&1 | tee /tmp/icu4c-newData-makeCheckEx.txt

# 10. Again, investigate each failure, fixing CLDR data or ICU test cases as
# appropriate, and repeating steps 4-7 and 9 until there are no errors.

# 11. Now with ICU4J, build and test without new data first, to verify that
# there are no pre-existing errors (or at least to have the pre-existing errors
# as a base for comparison):

cd $ICU4J_ROOT
ant all 2>&1 | tee /tmp/icu4j-oldData-antAll.txt
ant check 2>&1 | tee /tmp/icu4j-oldData-antCheck.txt

# 12. Now build the new data and test data for ICU4J

cd $ICU4C_DIR/source/data
make icu4j-data-install
cd $ICU4C_DIR/source/test/testdata
make icu4j-data-install

# 13. Now rebuild ICU4J with the new data and run tests:
# Keep a log so you can investigate the errors.

cd $ICU4J_ROOT
ant check 2>&1 | tee /tmp/icu4j-newData-antCheck.txt

# 14. Investigate test case failures; fix test cases and repeat from step 12,
# or fix CLDR data and repeat from step 4, as appropriate, until; there are no
# more failures in ICU4C or ICU4J (except failures that were present before you
# began testing the new CLDR data).

# 15. Check the file changes; then svn add or svn remove as necessary, and
# commit the changes.

cd $ICU4C_DIR/source
svn status
# add or remove as necessary

cd $ICU4J_ROOT
svn status
# add or remove as necessary

cd $HOME/icu/trunk/
# commit

# 16. For an official CLDR data integration into ICU, now tag the CLDR and
# ICU sources with an appropriate CLDR milestone (you can check previous
# tags for format), e.g.:

svn copy svn+ssh://unicode.org/repos/cldr/trunk \
svn+ssh://unicode.org/repos/cldr/tags/release-NNN \
--parents -m "cldrbug nnnn: tag cldr sources for NNN"

svn copy svn+ssh://source.icu-project.org/repos/icu/trunk \
svn+ssh://source.icu-project.org/repos/icu/tags/cldr-NNN \
--parents -m 'ticket:mmmm: tag the version used for integrating CLDR NNN'
Commit	Line	Data
f3c0d7a5 A	1	# Copyright (C) 2016 and later: Unicode, Inc. and others.
f3c0d7a5 A	2	# License & terms of use: http://www.unicode.org/copyright.html
b331163b	3	# Copyright (C) 2010-2014, International Business Machines Corporation and others.
729e4ab9 A	4	# All Rights Reserved.
	5	#
	6	# Commands for regenerating ICU4C locale data (.txt files) from CLDR.
	7	#
	8	# The process requires local copies of
	9	# - CLDR (the source of most of the data, and some Java tools)
57a6839d	10	# - ICU4J (used only for checking the converted data)
729e4ab9	11	# - ICU4C (the destination for the new data, and the source for some of it)
51004dcb A	12	# (Either check out ICU4C from Subversion, or download the additional
	13	# icu4c-*-data.zip file so that the icu/source/data/ directory is fully
	14	# populated.)
729e4ab9 A	15	#
	16	# For an official CLDR data integration into ICU, these should be clean, freshly
	17	# checked-out. For released CLDR sources, an alternative to checking out sources
	18	# for a given version is downloading the zipped sources for the common (core.zip)
	19	# and tools (tools.zip) directory subtrees from the Data column in
	20	# [http://cldr.unicode.org/index/downloads].
	21	#
	22	# The versions of each of these must match. Included with the release notes for
	23	# ICU is the version number and/or a CLDR svn tag name for the revision of CLDR
	24	# that was the source of the data for that release of ICU.
	25	#
	26	# Note: Some versions of the OpenJDK will not build the CLDR java utilities.
	27	# If you see compilation errors complaining about type incompatibilities with
	28	# functions on generic classes, try switching to the Sun JDK.
	29	#
	30	# Besides a standard JDK, the process also requires ant
	31	# (http://ant.apache.org/),
	32	# plus the xml-apis.jar from the Apache xalan package
	33	# (http://xml.apache.org/xalan-j/downloads.html).
	34	#
	35	# Note: Enough things can (and will) fail in this process that it is best to
	36	# run the commands separately from an interactive shell. They should all
	37	# copy and paste without problems.
	38	#
	39	# It is often useful to save logs of the output of many of the steps in this
	40	# process. The commands below put log files in /tmp; you may want to put them
	41	# somewhere else.
	42	#
	43	#----
	44	#
	45	# There are several environment variables that need to be defined.
	46	#
	47	# a) Java- and ant-related variables
	48	#
	49	# JAVA_HOME: Path to JDK (a directory, containing e.g. bin/java, bin/javac,
	50	# etc.); on many systems this can be set using
	51	# `/usr/libexec/java_home`.
	52	#
57a6839d	53	# ANT_OPTS: You may want to set:
729e4ab9	54	#
f3c0d7a5	55	# -Xmx3072m, to give Java more memory; otherwise it may run out
729e4ab9 A	56	# of heap.
729e4ab9 A	57	#
729e4ab9 A	58	# b) CLDR-related variables
	59	#
	60	# CLDR_DIR: Path to root of CLDR sources, below which are the common and
57a6839d A	61	# tools directories.
	62	# CLDR_CLASSES: Defined relative to CLDR_DIR. It only needs to be set if you
	63	# are not running ant jar for CLDR and have a non-default output
	64	# folder for cldr-tools classes.
729e4ab9	65	#
57a6839d A	66	# c) ICU-related variables
	67	# These variables only need to be set if you're directly reusing the
	68	# commands below.
729e4ab9 A	69	#
	70	# ICU4C_DIR: Path to root of ICU4C sources, below which is the source dir.
	71	#
	72	# ICU4J_ROOT: Path to root of ICU4J sources, below which is the main dir.
729e4ab9 A	73	#
	74	#----
	75	#
	76	# If you are adding or removing locales, or specific kinds of locale data,
	77	# there are some xml files in the ICU sources that need to be updated (these xml
	78	# files are used in addition to the CLDR files as inputs to the CLDR data build
	79	# process for ICU):
	80	#
	81	# icu/trunk/source/data/icu-config.xml - Update <locales> to add or remove
	82	# CLDR locales for inclusion in ICU. Update <paths> to prefer
	83	# alt forms for certain paths, or to exclude certain paths; note
	84	# that <paths> items can only have draft or alt attributes.
	85	#
51004dcb A	86	# Note that if a language-only locale (e.g. "de") is included in
	87	# <locales>, then all region sublocales for that language that
	88	# are present in CLDR data (e.g. "de_AT", "de_BE", "de_CH", etc.)
	89	# should also be included in <locales>, per PMC policy decision
	90	# 2012-05-02 (see http://bugs.icu-project.org/trac/ticket/9298).
	91	#
729e4ab9 A	92	# icu/trunk/source/data/build.xml - If you are adding or removing break
	93	# iterators, you need to update <fileset id="brkitr" ...> under
	94	# <target name="clean" ...> to clean the correct set of files.
	95	#
	96	# icu/trunk/source/data/xml/ - If you are adding a new locale, break
57a6839d A	97	# iterator, collation tailoring, or rule-based number formatter,
	98	# you may need to add a corresponding xml file in (respectively)
	99	# the main/, brkitr/, collation/, or rbnf/ subdirectory here.
729e4ab9 A	100	#
	101	#----
	102	#
	103	# For an official CLDR data integration into ICU, there are some additional
	104	# considerations:
	105	#
	106	# a) Don't commit anything in ICU sources (and possibly any changes in CLDR
	107	# sources, depending on their nature) until you have finished testing and
	108	# resolving build issues and test failures for both ICU4C and ICU4J.
	109	#
	110	# b) There are version numbers that may need manual updating in CLDR (other
	111	# version numbers get updated automatically, based on these):
	112	#
	113	# common/dtd/ldml.dtd - update cldrVersion
	114	# common/dtd/ldmlBCP47.dtd - update cldrVersion
	115	# common/dtd/ldmlSupplemental.dtd - update cldrVersion
	116	# tools/java/org/unicode/cldr/util/CLDRFile.java - update GEN_VERSION
	117	#
	118	# c) After everything is committed, you will need to tag the CLDR, ICU4J, and
	119	# ICU4C sources that ended up being used for the integration; see step 17
	120	# below.
	121	#
	122	################################################################################
	123
	124	# 1a. Java and ant variables, adjust for your system
729e4ab9 A	125
729e4ab9 A	126	export JAVA_HOME=`/usr/libexec/java_home`
f3c0d7a5	127	export ANT_OPTS="-Xmx3072m"
729e4ab9 A	128
	129	# 1b. CLDR variables, adjust for your setup; with cygwin it might be e.g.
	130	# CLDR_DIR=`cygpath -wp /build/cldr`
	131
	132	export CLDR_DIR=$HOME/cldr/trunk
729e4ab9 A	133	#export CLDR_CLASSES=$CLDR_DIR/tools/java/classes
	134
	135	# 1c. ICU variables
	136
f3c0d7a5 A	137	export ICU4C_DIR=$HOME/icu/trunk/icu4c
f3c0d7a5 A	138	export ICU4J_ROOT=$HOME/icu/trunk/icu4j
729e4ab9	139
57a6839d	140	# 2. Build the CLDR Java tools
729e4ab9 A	141
	142	cd $CLDR_DIR/tools/java
	143	ant jar
	144
57a6839d	145	# 3. Configure ICU4C, build and test without new data first, to verify that
729e4ab9 A	146	# there are no pre-existing errors (configure shown here for MacOSX, adjust
	147	# for your platform).
	148
	149	cd $ICU4C_DIR/source
	150	./runConfigureICU MacOSX
	151	make all 2>&1 \| tee /tmp/icu4c-oldData-makeAll.txt
	152	make check 2>&1 \| tee /tmp/icu4c-oldData-makeCheck.txt
	153
57a6839d	154	# 4. Build the new ICU4C data files; these include .txt files and .mk files.
729e4ab9 A	155	# These new files will replace whatever was already present in the ICU4C sources.
	156	# This process uses ant with ICU's data/build.xml and data/icu-config.xml to
	157	# operate (via CLDR's ant/CLDRConverterTool.java and ant/CLDRBuild.java) the
	158	# necessary CLDR tools including LDML2ICUConverter, ConvertTransforms, etc.
57a6839d	159	# This process will take several minutes.
729e4ab9 A	160	# Keep a log so you can investigate anything that looks suspicious.
	161
	162	cd $ICU4C_DIR/source/data
	163	ant clean
f3c0d7a5	164	ant all 2>&1 \| tee /tmp/cldr-newData-buildLog.txt
729e4ab9	165
57a6839d	166	# 5. Check which data files have modifications, which have been added or removed
729e4ab9 A	167	# (if there are no changes, you may not need to proceed further). Make sure the
	168	# list seems reasonable.
	169
	170	svn status
	171
57a6839d	172	# 6. Fix any errors, investigate any warnings. Some warnings are expected,
729e4ab9 A	173	# including warnings for missing versions in locale names which specify some
	174	# collationvariants, e.g.
	175	# [cldr-build] WARNING (ja_JP_TRADITIONAL): No version #??
	176	# [cldr-build] WARNING (zh_TW_STROKE): No version #??
	177	# and warnings for some empty collation bundles, e.g.
	178	# [cldr-build] WARNING (en): warning: No collations found. Bundle will ...
	179	# [cldr-build] WARNING (to): warning: No collations found. Bundle will ...
	180	#
	181	# Fixing may entail modifying CLDR source data or tools - for example,
57a6839d A	182	# updating the validSubLocales for collation data (file a bug if appropriate).
57a6839d A	183	# Repeat steps 4-5 until there are no build errors and no unexpected
729e4ab9 A	184	# warnings.
729e4ab9 A	185
57a6839d	186	# 7. Now rebuild ICU4C with the new data and run make check tests.
729e4ab9 A	187	# Again, keep a log so you can investigate the errors.
	188
	189	cd $ICU4C_DIR/source
	190	make check 2>&1 \| tee /tmp/icu4c-newData-makeCheck.txt
	191
57a6839d	192	# 8. Investigate each test case failure. The first run processing new CLDR data
729e4ab9	193	# from the Survey Tool can result in thousands of failures (in many cases, one
51004dcb	194	# CLDR data fix can resolve hundreds of test failures). If the error is caused
729e4ab9	195	# by bad CLDR data, then file a CLDR bug, fix the data, and regenerate from
57a6839d	196	# step 4. If the data is OK but the testcase needs to be updated because the
729e4ab9 A	197	# data has legitimately changed, then update the testcase. You will check in
729e4ab9 A	198	# the updated testcases along with the new ICU data at the end of this process.
51004dcb A	199	# Note that if the new data has any differences in structure, you will have to
	200	# update test/testdata/structLocale.txt or /tsutil/cldrtest/TestLocaleStructure
	201	# may fail.
57a6839d	202	# Repeat steps 4-7 until there are no errors.
729e4ab9	203
57a6839d	204	# 9. Now run the make check tests in exhaustive mode:
729e4ab9 A	205
	206	cd $ICU4C_DIR/source
	207	export INTLTEST_OPTS="-e"
	208	export CINTLTST_OPTS="-e"
	209	make check 2>&1 \| tee /tmp/icu4c-newData-makeCheckEx.txt
	210
57a6839d A	211	# 10. Again, investigate each failure, fixing CLDR data or ICU test cases as
57a6839d A	212	# appropriate, and repeating steps 4-7 and 9 until there are no errors.
729e4ab9	213
57a6839d	214	# 11. Now with ICU4J, build and test without new data first, to verify that
729e4ab9 A	215	# there are no pre-existing errors (or at least to have the pre-existing errors
	216	# as a base for comparison):
	217
	218	cd $ICU4J_ROOT
	219	ant all 2>&1 \| tee /tmp/icu4j-oldData-antAll.txt
	220	ant check 2>&1 \| tee /tmp/icu4j-oldData-antCheck.txt
	221
f3c0d7a5	222	# 12. Now build the new data and test data for ICU4J
729e4ab9 A	223
	224	cd $ICU4C_DIR/source/data
	225	make icu4j-data-install
f3c0d7a5 A	226	cd $ICU4C_DIR/source/test/testdata
f3c0d7a5 A	227	make icu4j-data-install
729e4ab9	228
57a6839d	229	# 13. Now rebuild ICU4J with the new data and run tests:
729e4ab9 A	230	# Keep a log so you can investigate the errors.
	231
	232	cd $ICU4J_ROOT
	233	ant check 2>&1 \| tee /tmp/icu4j-newData-antCheck.txt
	234
57a6839d A	235	# 14. Investigate test case failures; fix test cases and repeat from step 12,
57a6839d A	236	# or fix CLDR data and repeat from step 4, as appropriate, until; there are no
729e4ab9 A	237	# more failures in ICU4C or ICU4J (except failures that were present before you
	238	# began testing the new CLDR data).
	239
57a6839d	240	# 15. Check the file changes; then svn add or svn remove as necessary, and
729e4ab9 A	241	# commit the changes.
729e4ab9 A	242
57a6839d	243	cd $ICU4C_DIR/source
729e4ab9	244	svn status
f3c0d7a5	245	# add or remove as necessary
729e4ab9 A	246
	247	cd $ICU4J_ROOT
	248	svn status
f3c0d7a5 A	249	# add or remove as necessary
	250
	251	cd $HOME/icu/trunk/
	252	# commit
729e4ab9	253
f3c0d7a5 A	254	# 16. For an official CLDR data integration into ICU, now tag the CLDR and
f3c0d7a5 A	255	# ICU sources with an appropriate CLDR milestone (you can check previous
729e4ab9 A	256	# tags for format), e.g.:
	257
	258	svn copy svn+ssh://unicode.org/repos/cldr/trunk \
	259	svn+ssh://unicode.org/repos/cldr/tags/release-NNN \
	260	--parents -m "cldrbug nnnn: tag cldr sources for NNN"
	261
f3c0d7a5 A	262	svn copy svn+ssh://source.icu-project.org/repos/icu/trunk \
f3c0d7a5 A	263	svn+ssh://source.icu-project.org/repos/icu/tags/cldr-NNN \
729e4ab9 A	264	--parents -m 'ticket:mmmm: tag the version used for integrating CLDR NNN'
729e4ab9 A	265